How the Net Promotor Score (NPS) lost its way, despite Reichheld’s good intentions…

Join CX league

Receive weekly curated CX content by joining my newsletter with 1000+ readers from companies like Randstad, Forrester, Egon Zehnder, T-Mobile, Salesforce and more.

Let me start by saying that Reichheld to me is the author of my first bible, so this article has no intention to “bash” his work.

The goal is to help organisations understand why their CX is not delivering the results they expected and how the unintentional misuse of NPS might play a role in this.

There is not 1 company in the world, that does not know of or is actively using the Net Promotor Score in some form or other.

That’s an amazing achievement of Reichheld, together with Bain and Satmetrix.

The biggest achievement in my opinion is that they put the field of customer experience on the agenda as a top priority of many companies.

But… yes there is often a but… 😉

If all these companies across the globe have adopted the NPS since it’s launch in 2003, why is it that still to this day, only 7% of companies are leaders in the field of CX?

There are many reasons.

Though one of these reasons is that the NPS has lost its way, maybe even went on to take a life of its own, despite the good intentions Reichheld started with.

In the beginning…

In this HBR article from 2003, Reichheld introduces the Net Promotor Score.

His idea was ignited by a company (Rent-A-Car) that wanted to find a more easy measurement of loyalty.

Of to that point, measuring loyalty and its relation to profitability was a complex task with lengthy surveys about intentions to repurchase, word of mouth, price sensitivity, etc.

This company seemed to have found a simple way, by focusing only on two questions: the quality of the experience and the likelihood of renting again with Rent-A-Car.

And then they rated their branches on the % of customers that gave the highest experience rating.

Keep this in mind: the goal of this metric was to find out the most profitable customers.

To find a more simple way to measure loyalty and to predict growth.

The research of Reichheld

Reichheld was inspired by this and spend the next two years doing additional research to find out more and in the end found one metric to predict growth.

Hence, the Net Promotor Score was born.

In several industries it was not the best predictor, but in a lot of situations this one question did suffice.

And they found a lot of proof where companies with a high NPS score also showed very good growth figures.

So far so very good!

It seemed to be a solution to the complex measurement of loyalty, which was the field of Reichheld’s focus from the beginning (if you haven’t read his 1996 book The Loyalty Effect yet, you definitely should).

Application of NPS with direct feedback

Enterprise Rent-A-Car used their new found two metrics as follows.

Interviewers contacted the customers and asked them the two questions.

As soon as the customer gave a neutral or negative score, the customer was asked permission to immediately forward this information to the branch manager.

The branch manager was then trained to apologize, find the root cause and fix it.

A highly individual approach, only impacting the customers that were interviewed ánd that gave neutral of negative scores.

This is the second thing you need to remember, next to the first thing that the goal was to find a metric that predicts profitability.

Misread title blog and book

Although the original blog title “The One Number You Need To Grow” and the original book title “The Ultimate Question” seem to indicate otherwise, Reichheld himself is the first to say you always need a good set of metrics and not just 1 question to manage the (customer) experience of your company.

But the ease of the metric and maybe also the marketing behind it, may have seduced organisations to use NPS as it was not originally meant to be used.

Apart from methodological issues that have been raised about NPS in the last few years, there are also more practical issues:

  1. Direct feedback is only impacting a very small number of customers;
  2. Solving individual customer’s issues virtually impossible for large companies;
  3. Finding the root causes at scale was never the intention of the NPS methodology
1. Direct feedback is only impacting a very small number of customers

Calculate with me for a minute.

Let’s say you send 100% of your customers a survey with the NPS question.

Let’s say that 20% of your customer respond to this survey.

Let’s say that 25% of these customers rate you a detractor score (0-6).

Let’s say that 50% of these detractors indicate they are ok with you contacting them.

Let’s say you have the capacity to contact 100% of these detractors.

100% * 20% * 25% * 50% = 2,5% of your customers that give you a chance to turn them from detractors into passives or promotors.

97,5% of your customers are completely not impacted by this approach.

And the percentages I use here are very high, so in reality this 2,5% is probably even lower.

But you get the point I’m trying to make.

2. Solving individual customer’s issues virtually impossible for large companies

Even in the situation above, if you are a large company with 2 million customers, you would have to contact 2,5% * 2.000.000 = 5.000 customers.

Let alone if you don’t want to have an impact on just 2,5% of your customers but all of them.

That would mean you need interviewers that are able to contact 100% of your customers.

Let’s say again 25% of them is not happy and 100% is ok to be contacted to solve the issue.

In this case you would have to contact 500.000 customers.

And this is a very very very hypothetical case.

The point I’m obviously trying to make, is that you can not use direct feedback based NPS in very large organisations.

In smaller B-to-B settings this may work fine, but in B-to-C it’s definitely not an option.

This is something that many organisations of course had already found out, and that’s why the root cause analysis became part of the NPS methodology.

So now the NPS methodology is no longer just the NPS question and then follow up with detractors.

Now the NPS methodology consists of the NPS questions and then either a predefined selection list of rootcauses or an open text asking “why this score?”.

3. Finding the root causes at scale was never the intention of the NPS methodology

Quick check in: remember that the goal of the NPS was to find a metric that could predict profitability? And not to find the drivers of a great experience?

Organisations started to move away from contacting the customers that were unhappy (so-called direct feedback), to discovering the cause of detractors on a larger scale.

If they are still contacting detractors, it’s often a small part of them, for example the most valuable customers in a business to business setting for example).

If I ignore the option of root causes from a selection list (with all the obvious internal bias), then still finding the root causes with open text is an issue.

Even if you nowadays have smart artificial intelligence that can analyse all these open answers.

These are the 3 main issues:
  1. You are finding a list of rational explanations in the text, while we as human beings are far from rational people. A big risk of a mismatch between the two with a focus on improving the wrong drivers.
  2. You have no idea whether you are on the right track. Are the drivers you have found the ones that matter the most to your customers? Or are you missing some crucial drivers?
  3. You don’t now if root cause number 1 is 2, 5 or 10 times more relevant for their experience then one of the other root causes.

If you read through these three issues, you see that the common denominator is lack of direction.

You are not in the driving seat of customer experience.

You are lacking hard steerable insights on what matters the most to your customers.

And this is the explanation why, even though the whole world has practically adopted the NPS methodology, so many organisations (93%) are still 17 years later struggling to be CX leaders.

The good news?

There are years and years of valuable statistical methodologies that can fix this problem easily.

By combining those statistics to the context of customer journeys, you know exactly where to make the difference and see measurable results within a few months time.

Not only creating happy customers, but also bring your employees back to their purpose of why the originally joined your organisation.

Join CX league

Receive weekly curated CX content by joining my newsletter with 1000+ readers from companies like Randstad, Forrester, Egon Zehnder, T-Mobile, Salesforce and more.

More To Explore

CX transformatie

Podcast: Interview met Danny Peters (Milkymap)

Vandaag duiken we in een onderwerp dat de laatste tijd veel aandacht krijgt: journey management. Zoals ik al in eerdere podcasts heb vermeld, wordt journey management steeds belangrijker. En wie kan daar beter over vertellen dan Danny? We kennen elkaar al een tijdje, hoewel niet heel goed, maar genoeg om te weten dat zijn inzichten

CX transformation

Podcast: Interview with Danny Peters (Milkymap)

Today’s topic is something that’s been gaining a lot of traction lately: journey management. As mentioned in previous podcasts, journey management is becoming increasingly important, and I couldn’t think of a better person to discuss this with than Danny. We’ve known each other for quite some time, although not super well, but enough to know